Vehicle electronics can’t roll past him

Don Mersino is a test engineer for Magna Electronics (Holly, MI), a maker of automotive-control systems that include mirror electronics that provide outside air temperature, compass direction, and auto dimming features; control modules for power-lift gates, deployable running boards, and variable blower controllers; and automotive camera systems. Mersino recently discussed his work with senior technical editor Martin Rowe.

Q: What is your role in test?

A: I’m responsible for complete functional testing of production products. I develop PC-based testers that verify operation of automotive controls. I handle testing from the time we quote on new business through training production technicians.

Q: How do you get the specifications on I/O signals for each product?

A: Our hardware and software engineers work with customers to get details on a product’s input and output signals. I receive signal specifications from the designers and develop an automated test system.

Q: What do you use to generate input signals for a controller, and what do you use to check a con-troller’s outputs?

A: We use PCI multifunction I/O cards. The cards connect to an interface board that contains relays to simulate discrete inputs. We often need to add more digital I/O lines than a multifunction card can provide. We usually add another multifunction card, but I’ve been experimenting with a USB digital I/O module. Counter/timers simulate pulse-width modulated and Hall-effect signals.

I also develop machine-vision systems to verify LEDs and displays. Finally, I use frame grabbers to focus and test the camera products.

Q: How does the tester connect to the UUT?

A: A ribbon cable from a multifunction card connects to a custom interface board that breaks the connections out to screw-terminal connectors. Wires from the screw-terminal connectors attach to any necessary loads and to quick-disconnect connectors as part of a wire harness. The wire harness connects to another set of quick-disconnect connectors on a test fixture. Pogo pins on pneumatic cylinders connect test signals and loads to the UUT.

Q: What’s the process for developing a tester?

A: After I have an understanding of the I/O signals, I assemble the I/O cards and interface boards. About 85% of the work involves software. I look for a tester with software that most closely matches my needs for a new system, then I modify the code and user interface, which is written in Visual Basic 6. For the most part, I can copy the data-handling routines from one system to another.

Q: What do your data-handling routines do?

A: The data-handling routines store measurements in a database on the PC, and we upload data to the network. Data is available through a Web server. We use the data to track production yields and throughput. The data also helps us with field returns. For example, if a product fails in a way that a production tester should have caught, then the customer usually wants to talk to me or another test engineer to learn how we tested the product.

Q: How do you introduce a tester to production?

A: I work with the manufacturing engineers to secure a place for the tester on the production floor. I then show the tester and UUT schematics to technicians and operators. I explain how the product works and what the tester does. I also show test operators how to load the UUT onto the tester and how to label a product after its test.

Q: Do you use the testers for production only or do you also use them for design verification?

A: Mostly, we use the testers for production. Once in a while, a design engineer will use a system to check a modified product.